HDFS is agnostic about the contents of the data you store. Think about it: Line ending character is not the universal way for files to separate their records.
This question's been asked several times before (search on http://search-hadoop.com for example). Read http://wiki.apache.org/hadoop/HadoopMapReduce to understand how despite HDFS splitting at exactly the 64th megabyte, MR (or other HDFS file reading operations) make sure to read records whole. On Sat, Jan 4, 2014 at 3:59 PM, VJ Shalish <vjshal...@gmail.com> wrote: > Hi, > > While creating the blocks for a file containing n number of lines, how does > Hadoop take care of the problem of not Cutting a line in between while > creating blocks? > Is it taken care of by Hadoop? > > Thanks > Shalish. -- Harsh J
